Update README.md
Browse files
README.md
CHANGED
|
@@ -3,6 +3,8 @@ language:
|
|
| 3 |
- dv
|
| 4 |
base_model:
|
| 5 |
- openai-community/gpt2
|
|
|
|
|
|
|
| 6 |
---
|
| 7 |
|
| 8 |
# GPT 2 DV base
|
|
@@ -15,7 +17,6 @@ This is a GPT-2 model fine-tuned on Dhivehi language texts. The model was traine
|
|
| 15 |
- **Language:** Dhivehi (ދިވެހި)
|
| 16 |
- **Training Data:** Dhivehi Wikipedia articles
|
| 17 |
- **Last Updated:** 2024-11-25
|
| 18 |
-
- **License:** MIT
|
| 19 |
|
| 20 |
## Performance Metrics
|
| 21 |
|
|
@@ -58,7 +59,6 @@ print(generated_text)
|
|
| 58 |
The model was trained using the following configuration:
|
| 59 |
- Base model: GPT-2
|
| 60 |
- Training type: Full fine-tuning
|
| 61 |
-
- Hardware: NVIDIA A40 GPU
|
| 62 |
- Mixed precision: FP16
|
| 63 |
- Gradient checkpointing: Enabled
|
| 64 |
|
|
@@ -88,14 +88,4 @@ This model is suitable for:
|
|
| 88 |
Not intended for:
|
| 89 |
- Critical or production systems
|
| 90 |
- Decision-making applications
|
| 91 |
-
- Tasks requiring factual accuracy
|
| 92 |
-
|
| 93 |
-
## Citation
|
| 94 |
-
|
| 95 |
-
```bibtex
|
| 96 |
-
@misc{dhivehi-gpt2,
|
| 97 |
-
title = {Dhivehi GPT-2: A Language Model for Dhivehi Text Generation},
|
| 98 |
-
year = {2024},
|
| 99 |
-
publisher = {Hugging Face},
|
| 100 |
-
}
|
| 101 |
-
```
|
|
|
|
| 3 |
- dv
|
| 4 |
base_model:
|
| 5 |
- openai-community/gpt2
|
| 6 |
+
datasets:
|
| 7 |
+
- wikimedia/wikipedia
|
| 8 |
---
|
| 9 |
|
| 10 |
# GPT 2 DV base
|
|
|
|
| 17 |
- **Language:** Dhivehi (ދިވެހި)
|
| 18 |
- **Training Data:** Dhivehi Wikipedia articles
|
| 19 |
- **Last Updated:** 2024-11-25
|
|
|
|
| 20 |
|
| 21 |
## Performance Metrics
|
| 22 |
|
|
|
|
| 59 |
The model was trained using the following configuration:
|
| 60 |
- Base model: GPT-2
|
| 61 |
- Training type: Full fine-tuning
|
|
|
|
| 62 |
- Mixed precision: FP16
|
| 63 |
- Gradient checkpointing: Enabled
|
| 64 |
|
|
|
|
| 88 |
Not intended for:
|
| 89 |
- Critical or production systems
|
| 90 |
- Decision-making applications
|
| 91 |
+
- Tasks requiring factual accuracy
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|